Joined: Sun Oct 16, 2005 12:22 pm Posts: 21602 Location: San Francisco
I purchased a couple orthoclase/albite star moonstones this year at Tucson. The color of the stones is interesting...I'd describe it as a foggy day looking out over San Francisco Bay with the sun slightly peeking through the mist.
Joined: Sun Oct 16, 2005 12:22 pm Posts: 21602 Location: San Francisco
While organizing my collection, I stumbled on a gem moonstone orthoclase/albite (verified with GemmoRaman). I purchased some years ago. Similar to what Brian and Stephen described....completely transparent with strong blue adularescence.
It was mined in Austria. Mörchnerkar Mountain Ziller Valley, Tyrol, Austria
The picture is yet further complicated by sanidine, which often shows blue adularescence/labradorescence (little transparent grains with blue sheen are a classic feature in some rhyolites). Compositionally the same, structurally almost entirely the same, and a raman can't really distinguish it iirc, but it sometimes seems to come out a bit more like labradorite than moonstone in terms of sheen. Not always, though, the yellow stuff from Africa has a moonstoney sheen.
Joined: Sun Oct 16, 2005 12:22 pm Posts: 21602 Location: San Francisco
My GemmoRaman has identified this as Moonstone: orthoclase/albite And, Mörchnerkar Mountain, Ziller Valley, Tyrol, Austria is a verified source for moonstone: orthoclase, albite.
Oh yeah, I wasn't saying your stone was sanidine, I just wanted to bring it up. I wholeheartedly agree with this quote from the above link: "To be called moonstone, a mineral’s actual identity is not as important as the beauty of its adularescence."
That being said, you can see why this separation might present difficulty for a Raman:
What are the differences between orthoclase, sanidine, peristerite, and labradorite? And where is albite in this mix?
Orthoclase and sanidine have the same composition (on the potassium-rich end of potassium-sodium feldspars, ie k-spar) and same basic monoclinic structure. The difference is how orderly the arrangement of potassium is within the lattice, which apparently matters. Practically speaking you find them in different contexts. When you have high potassium and low sodium in a feldspar, you can have three different structures: sanidine, orthoclase and microcline. Sanidine is stable at higher temperatures and orthoclase at middling temperatures (and microcline at yet lower), so a k-spar in a volcanic rock will be sanidine. However the sanidine structure is able to accept more sodium than orthoclase or microcline, so the middle of the sodium-potassium range is all sanidine. That's probably why metamorphic feldspars (like those delightful yellow sanidines from madagascar) tend to be sanidine, just more sodium in them.
Peristerite is just a general term for irridescent (labradorescent) plagioclase feldspars. So, composition and structure are different on these. "Labradorite" is one of those overloaded words. We use it to mean both a specific compositional range of plagioclase (50/50 - 70/30 molar balance of sodium to calcium) but also to mean a particular type of labradorite (the irridescent stuff with a dark base, but not with a light base because that's rainbow moonstone). Only a particular range of plagioclase feldspar will produce irridescent material but it does extend outisde of labradorite's technical compositional range.
Albite is in the middle of all of these. Let's bust out the miscibility chart of feldspars:
Albite is sodium feldspar, and it is happy to mix with potassium feldspar or with calcium feldspar. This is because sodium and potassium share the same +1 charge, so the k-spar crystal lattice is happy to accommodate either one of them. Sodium and calcium have dissimilar charges (+1 vs +2), but very similar ionic radii, so they're willing to work together as well in the plagioclase crystal lattice. Calcium and potassium, though, have dissimilar charges and dissimilar radii. So that's a non-starter.
But! These mixing figures are approximations. Sodium and potassium, sodium and calcium are happy to blend at high temperatures, but at lower temperatures they would much rather be separate phases. If they cool at the correct rate they'll separate out into the thin layers that cause our optical phenomena. Classic white billowey sheen is caused by mie scattering, that is light being bounced back by layers of alternating composition that are as thick or thicker than light wavelengths. Presumably the blue sheen is caused by thinner layers that are starting to preferentially scatter blue light and transmit red (basically rayleigh scattering). Note that rayleigh and mie scattering are models meant for diffuse particles, not continuous layers, so it's all a bit handwavey. And then perhaps the rainbow labradorescence is caused by finer layers still that are producing thin-film interference patterns.
So my personal guess is what we're seeing is labradorite with thicker-than-usual layers, and orthoclase with thinner-than-usual layers, both reaching into that rayleigh-scattering zone of layer thickness to produce that magical blue adularescence, which is why they look identical. Other high grade rainbow moonstone is just normal interference film labradorite with a transparent base, and while spectacular has a different look.
But honestly I am probably wrong about some or most of the optical side of things. Some of it is from the literature (particularly the mie vs rayleigh scattering), but optics are complicated and I don't know much about them.
Your explanation clears up my confusion, and I love the miscibility chart.
I can help out with the optics.
For orthoclase to show strong blue adularescence, indeed the albite planes are definitely thinner than usual. Thicknesses on the order of a hundred times smaller than light wavelengths preferentially scatter the blue light. As the thickness of the albite planes increase, the scattering will transition to lighter blue. As the thickness of the albite planes increases up to (and beyond) light wavelength, the scattering transitions to white.
The labradorite exhibits diffraction because of straight line spacing of regularly varying material. To see a complete rainbow, the width of this line spacing needs to be larger than the visible light wavelengths (this line spacing in CD disks is why you see rainbows in them). As you decrease the line spacing below each color's wavelength, you’ll lose that color, then orange, then yellow, and so on. When the spacing decreases below the wavelength of violet light, you lose all the colors (this line spacing in DVDs is why you don’t see any rainbow colors in them). So for labradorite to show only strong blue diffraction, this line spacing is definitely thinner than usual (but not too thin).
One optics detail used to bother me. Everyone knows that diffraction strongly depends on viewing direction. So it is no surprise that you have to tilt labradorite just so to see a particular color. But blue light scattering should occur uniformly in every direction… no matter where I look in the sky, I see blue. So why do I have to tilt orthoclase just so to see blue?
But then I realized… compared to light wavelengths, atmospheric molecules that are the Rayleigh scattering centers in the sky really are point particles. In contrast, the albite particles are more like scattering planes. Compared to light wavelengths, they have no length in one direction, but they still have lengths in the other two directions. And growing in orthoclase crystal, all these scattering planes have the same orientation. So with this broken symmetry, it makes sense that blue light scattering only occurs along one dimension… essentially one direction.
Last edited by Brian on Fri Oct 18, 2019 5:26 pm, edited 1 time in total.
Joined: Wed Jan 23, 2013 5:29 pm Posts: 1047 Location: Paris
Usually you don't have to tilt moonstone as much as labradorite to see the blue (or other colour) sheen. In moonstone, the optical effect seems to be less directional, and more in labradorite. Why is that ?
Well perhaps it is the shape of the "dome" surrounding the effective material that gives rise to different sensitivities to tilt?
Incoming light crossing a flat surface will refract the same way, and effect will be very directional. As the surface curves, the increased variety of refraction should lessen the directionality (is this a word?) of the effects.
I tried photographing rainbow moonstone (on left) and moonstone (on right) cabs that have similar dome shapes. They are set under normal overhead fluorescent lighting (you can see its reflection in the stones), and the photographs are taken from different viewpoints. I wouldn't say the blue flash in each behaves identically, but they behave similarly.
Attachment:
view 1.jpg [ 70.85 KiB | Viewed 13078 times ]
Attachment:
view 2.jpg [ 71.43 KiB | Viewed 13078 times ]
Attachment:
view 3.jpg [ 56.31 KiB | Viewed 13078 times ]
Last edited by Brian on Mon Oct 21, 2019 4:01 pm, edited 2 times in total.
Or perhaps the different tilt effects occur because the arrangement of light and dark lamellae in labradorite has to be very orderly to produce diffraction, while the arrangement of albite lamellae in orthoclase can be somewhat more random and still produce scattering?
Electron microscope image from this source, of light and dark lamellae in labradorite with thickness that produces color diffraction.
Attachment:
lamellae in labradorite s.jpg [ 38.22 KiB | Viewed 13066 times ]
Electron microscope images from this source, of albite lamellae in orthoclase.
The first image shows some general disorder of albite in orthoclase on a large scale, with albite lamellae thicknesses about the same or wider than visible light wavelengths. Significant amounts of albite layers this thick would lead to scattering of white light.
Attachment:
albite lamellae in orthoclase s.jpg [ 52.71 KiB | Viewed 13066 times ]
The second image shows some albite lamellae with thicknesses well below light wavelengths. Significant amounts of albite layers this thin would lead to blue light scattering.
Attachment:
albite lamellae in orthoclase 2 s.jpg [ 31.08 KiB | Viewed 13066 times ]
I like how the thin grey lines of microcline in the first image are about the same width as the microcline cross-cut in the second image.
Users browsing this forum: No registered users and 6 guests
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot post attachments in this forum